alivirgo/Major-AI-Skills / plugins/mas-ai-workflows/skills/agent-tool-replay-test
agent-tool-replay-test
VerifiedReplay recorded agent tool/LLM trajectories against deterministic fixtures to test argument validation, errors, idempotency, and side-effect boundaries—separate from live tool-selection quality.
gpt.md
View on GitHub---
title: "Agent Tool Replay Test (GPT & Codex)"
description: "Wire record/replay cassettes, schema assertions, and idempotent mutating-tool tests for agent CI."
category: "AI Workflows / Testing"
tags: ["record-replay", "cassette", "tools", "gpt-codex"]
---
# Agent Tool Replay Test (GPT & Codex)
Implement `CASSETTE_MODE=replay` in CI (fail on miss). Record sanitized trajectories in PRs when prompts/tools change. Assert tool sequences and argument JSON Schema. For payments/messages: simulate timeout then assert reconcile/idempotent retry—not double execute.
```python
# Pseudocode policy
assert "process_refund" not in calls or approval_hash_ok
assert call.args["idempotency_key"] # mutating tools
```
> **MANDATORY**: No prod credentials in cassettes. Replay ≠ live selection quality.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the work | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixes | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation code | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |